Epothilones A and B represent a new class of microtubuli-stabilising cytotoxic active ingredients (see Gerth, K. et al., J. Antibiot. 49, 560-3 (1966)) of the formulae: wherein R signifies hydrogen (epothilone A) or methyl (epothilone B).
Since the description of these epothilones (see WO 93/10121), several methods have been disclosed for the synthesis and manufacture both of the epothilones and mainly of numerous epothilone derivatives (collectively called “epothilones” hereinafter), for example those disclosed in WO 99/03848, WO 00/49020, WO 00/49021, WO 00/47584, WO 00/00485, WO 00/23452, WO 99/03848, WO 00/49019, WO 99/07692, WO 98/22461, WO 99/65913, WO 98/38192, WO 00/50423, WO 00/22139, WO 99/58534, WO 97/19086, WO 98/25929, WO 99/67252, WO 99/67253, WO 00/31247, WO 99/42602, WO 99/28324, WO 00/50423, WO 00/39276, WO 99/27890, WO 99/54319, WO 99/54318, WO 99/02514, WO 99/59985, WO 00/37473, WO 98/08849, U.S. Pat. No. 6,043,372, U.S. Pat. No. 5,969,145, WO 99/40047, WO 99/01124, and WO 99/43653. In addition to epothilone A and epothilone B, especially the epothilones D and E, described in WO 97/19086 and WO 98/22461, and the epothilones E and F, described in WO 98/22461, as well as the epothilones described in WO 99/02514 show interesting properties.
As an example for the therapeutic use, International Application WO 99/43320 describes a number of ways of administration of epothilones as agents against proliferative diseases, especially tumor diseases, that, due to their similar mechanism, act in a way comparable to that of Taxol®, a well-known and marketed anticancer agent. WO 99/39694 discloses some specific formulations of the epothilones, especially A and B.
The epothilones, especially epothilone A and most preferably epothilone B, offer a number of advantages in comparison to established treatments, especially also in cases where tumors have become refractory to the treatment with Taxol. Therefore, methods for their synthesis in larger amounts are urgently required in order to meet anticipated demands.
The most efficient production processes so far at least comprise some biosynthetic steps and isolation of epothilones from culture media or the like.
Originally, the extraction of natural substances by means of myxobacteria, especially the epothilones from the cell strain Sorangium Cellulosum Soce90 (deposited under no. 6773 at the German Collection of Microorganisms, see WO 93/10121) was described in literature. In order to obtain a satisfactory concentration of the natural substances, especially the epothilones, previously an adsorbate resin based on polystyrene was always added to the culture medium for absorption to the medium, for example Amberlite XAD-1180 (Rohm & Haas, Frankfurt, Germany).
However, the disadvantage of this process is that, on a large scale, it leads to an abundance of problems. Valves are impaired by the globules of resin, pipes can block, and apparatus may be subject to greater wear due to mechanical friction. The globules of resin are porous and therefore have a large inner surface area (about 825 m2/gram resin). Sterilisation becomes a problem, as air enclosed in the resin is not autoclaved. Thus, the process cannot be practicably carried out on a large scale using resin addition during cultivation of the microorganisms that produce epothilones.
Therefore an advanced process for the production of epothilones, especially of epothilones A and B, was found and described in WO 99/42602. That method comprises complexing of epothilones from culture media of epothilone-producing microorganisms, said media comprising cyclodextrines or other complex-forming agents, mixing of the cell-free culture medium (e.g. filtrate or centrifugate of said culture medium) with a synthetic resin, for example a resin based on styrene/divinylbenzene copolymers as matrix, such as Amberlite XAD-16 (Rohm & Haas Germany GmbH, Frankfurt, Germany) or Diaion HP-20 (Resindion S.R.L., Mitsubishi Chemical Co., Milan, Italy) in order to absorb the epothilones and desorption, especially with an alcohol, most preferably isopropanol. This is followed by addition of water to the alcohol phase, removal of the solvent phase (preferably by evaporation), phase separation of the resulting residue in the presence of an ester, especially ethyl acetate or isopropyl acetate, usually molecular filtration (gel chromatography) of the dried ester phase, separation of the resulting epothilone mixture by reverse phase HPLC (preferably by elution with a mixture of nitrile/water, e.g. acetonitrile/water), and optionally further purification by phase separation in the presence of a water/ether mixture, preferably subsequent adsorption chromatography on silica gel in order to achieve further removal of impurities, and crystallisation/recrystallisation.
Though a useful progress and appropriate for industrial scale production, this method still suffers from certain disadvantages.
For example, in order to obtain sufficient purity, it is advisable to make use of either the molecular filtration step or the silica gel adsorption chromatography step or both. More difficulties come from the phase separation in the presence of an ester, such as ethyl acetate, which (especially due to the long time for phase separation of the water/ester phase in large industrial scale) is very time-consuming, as well as the subsequent evaporation, which in addition is difficult to handle in view of foaming and sputtering.
It is thus a problem to be solved by the present invention to avoid as many of the above difficulties as possible and find new and advantageous ways for the isolation of epothilones, especially epothilones A and B, after their adsorption to a resin.